Exploring implicit value frames in media through natural language processing

Partners: Delft University of Technology & University of Amsterdam 
Type: Postdoc 
Duration: 2024—2027

Values represent the fundamental motivations that drive opinions and actions, encompassing concepts such as fairness and self-determination. They function as a lens through which media is both created and interpreted, guiding writers in framing news items and readers in their reactions. However, due to their abstract nature, values often remain implicit and are not explicitly articulated in everyday discourse. This study aims to uncover the latent value-laden language within media narratives to enhance understanding between media producers and consumers. Utilizing advanced natural language processing (NLP) techniques and state-of-the-art language models, the goal of the project is to analyse how media outlets frame core values such as justice, equality, freedom, and security. The research seeks to evaluate how these value frames influence public perception and shape social behavior. By revealing how news coverage subtly reinforces, challenges, or reshapes collective values, this project will contribute to a deeper understanding of the media's role in influencing cultural and moral landscapes in an increasingly polarized world.